Multiscale Modeling of Blood Flow and Platelet Mediated Thrombosis

 描述（由申请人提供）：心血管疾病仍然是发达国家的主要死因，每年占全球所有死亡的近30%，在美国占35%。在美国，冠状动脉疾病（CAD）及其相关血栓形成风险是导致6例死亡的原因之一。巧合的是，为严重心血管疾病患者提供救生解决方案的植入式血液再循环装置也面临血栓形成和血栓栓塞并发症，需要复杂的终身抗凝治疗。血管疾病过程和器械相关血栓性并发症的潜在机制相互交织。血管疾病中的血栓形成通过血液成分与受损血管壁的相互作用而增强，并且在心血管病理中产生的非生理性流动模式通过慢性激活血小板来启动和增强止血反应。同样，器械血栓形成是由病理性流场和与异物表面接触引起的。血小板活化后会发生复杂的生物化学和形态学变化。分子水平事件和宏观输运之间的不同时空尺度的耦合是一个主要的建模和计算挑战，这需要多学科集成的多尺度数值方法。连续体方法在其覆盖较小分子机制（例如血小板活化期间丝状伪足形成）的能力方面是有限的。利用分子动力学（MD）来覆盖所涉及的多尺度是计算上禁止的。在这个应用程序中，我们提供 发展一个全面的国家的最先进的多尺度数值方法，将能够弥合之间的差距差距的宏观运输和随后的分子事件。我们将使用一种综合的耗散粒子动力学（DPD）和粗粒分子动力学（CGMD）方法，该方法允许血小板连续改变其形状，并通过生物力学转导连接链协同激活，与其他血液成分和凝血因子相互作用，聚集，相互作用并粘附于血管和器械。在该多尺度模型中，将机械转导CGMD底部血小板活化模型嵌入到DPD血流顶部模型中。宏观尺度模型的动态应力将交互地转换为血小板内相关细胞内事件的微观到纳米尺度模型。该模型的预测将在一组精心设计的实验中进行体外验证。这将根据以下具体目标实现：我们将开发血小板介导的血栓形成的机械转导模型，其中使用DPD在μ m长度和ms时间尺度上进行流动诱导血栓形成的顶部/宏观尺度模型，其中多个流动的血小板彼此相互作用并与血管壁或器械相互作用，将与使用CGMD的底部/微尺度模型在nm-长度和ps-时间尺度上完全耦合，其中具有多种细胞内成分的血小板在活化期间随着血小板失去其静止盘状形状和丝状伪足生长而演变。顶部和底部模型将相互连接，使得血液动力学将对激活后的血小板形状变化和血小板聚集以及血栓形成做出交互反应。还将模拟通过抗血小板剂调节血小板机械性质的影响。所有模型方面都将在体外在一系列精心设计的实验中进行验证，这些实验表征血小板的机械特性，并使用血流实验，其中将复制导致流动诱导的血小板活化的条件，以及血小板壁和血小板膜的粘附实验。 将测量血小板与装置的相互作用，并将用调节剂对血小板进行预处理。这些数据将被用来微调大量的模型参数参与这个多尺度模拟和验证模型预测。独立的第三方评估模型的可信度也包括作为该项目的一个组成部分。e还将专注于开发适用于超可扩展大型HPC集群的高效算法，以降低高昂的计算成本，从而将这种雄心勃勃的大型多尺度模拟带入多尺度建模社区，并使其能够适应其他相关建模需求和兴趣。为了进一步实现这些技术，将建立大型可共享的数据库，其中将存放软件工具、数字代码、模型和实验数据以及协议，并将提供使用它们的指导。该项目的领导者将积极参与各种MSM联盟工作组，以进一步传播项目成果，并与建模社区分享。所提出的方法代表了一个范式的转变，在新兴领域的多尺度模拟及其应用，以解决复杂的临床问题，在工程和生物学的接口。预测循环条件下动脉血栓形成的进展，提供与现有的基于药物的治疗相比改进的药物管理的工具，以及提供用于开发具有降低的促凝性的下一代装置的建模工具，可以导致降低的死亡率，改善的患者的生活质量，以及随之而来的医疗保健费用的经济负担的总体降低。

 DESCRIPTION (provided by applicant): Cardiovascular diseases remain the leading cause of death in the developed world, accounting for near 30% of all deaths globally and 35% in the US annually. Coronary artery disease (CAD) with its associated thrombotic risk is responsible for 1 of 6 deaths in the US. Coincidentally, implantable blood recirculating devices, which have provided lifesaving solutions to patients with severe cardiovascular diseases, are burdened with thrombosis and thromboembolic complications, mandating complex life-long anticoagulation. The mechanisms underlying vascular disease processes and device-related thrombotic complications are intertwined. Thrombosis in vascular disease is potentiated by the interaction of blood constituents with an injured vascular wall and the non-physiologic flow patterns generated in cardiovascular pathologies initiate and enhance the hemostatic response by chronically activating the platelets. Similarly, device thrombogenicity is induced by pathological flow fields and contact with foreign surfaces. Upon activation platelets undergo complex biochemical and morphological changes. The coupling of the disparate spatio-temporal scales between molecular level events and the macroscopic transport represents a major modeling and computational challenge, which requires a multidisciplinary integrated multiscale numerical approach. Continuum approaches are limited in their ability to cover the smaller molecular mechanisms such as filopodia formation during platelet activation. Utilizing molecular dynamics (MD) to cover the multiscales involved is computationally prohibitive. In this application we offer to develop a comprehensive state-of-the-art multiscale numerical methodology that will be able to bridge the gap between the macroscopic transport and the ensuing molecular events. We will use an integrated Dissipative Particle Dynamics (DPD) and Coarse Grained Molecular Dynamics (CGMD) approach that allows platelets to continuously change their shape and synergistically activate by a biomechanical transductive linkage chain, interact with other blood constituents and clotting factors, aggregate, and interact and adhere to the blood vessels and devices. In this multiscale model, a mechanotransduction CGMD bottom platelet activation model is embedded into a DPD blood flow top model. The dynamic stresses of the macroscale model will be interactively translated to the micro to nanoscale model of the intra-platelet associated intracellular events. The model predictions will be validated in vitro in a carefully designed set of experiments. This will be achieved according to the following specific aims: We will develop a mechanotransduction model of platelet mediated thrombosis where a top/macro-scale model of flow-induced thrombogenicity using DPD at the µm-length and ms-time scales, in which multiple flowing platelets interact with each other and blood vessel walls or devices, will be fully coupled with a bottom/micro-scale model using CGMD at the nm-length and ps-time scales, in which platelets with multiple intracellular constituents evolve during activation as platelet lose their quiescent discoid shape and filopodia grow. The top and bottom models will be interfaced such that the hemodynamics will interactively respond to platelet shape change upon activation and platelet aggregation and thrombus is formed. The effect of modulating platelet mechanical properties via antiplatelet agents will be modeled as well. All model aspects will be validated in vitro in a series of carefully designed experiments characterizing the mechanical properties of platelets and using blood flow experiments where conditions leading to flow induced platelet activation will be replicated, as well as experiments where platelet-wall and platelet device interactions will be measured and where platelets will be pretreated with modulating agents. These data will be used to fine tune the large number of model parameters involved in this multiscale simulation and for validating the model predictions. An independent 3rd party evaluation of the model credibility is also included as an integral part of the project. e will also concentrate on the development of efficient algorithms adapted for ultra-scalable large HPC clusters to reduce prohibitive computation costs, so as to bring such ambitious large multiscale simulations within the reach of the multiscale modeling community at large and enable to adopt it to other relevant modeling needs and interests. To further enable these technologies, large sharable data base will be created where software tools, numerical codes, model and experimental data and protocols will be deposited and guidance will be provided for using them. The leaders of the project will be active in various MSM consortium working groups to further disseminate the project outcomes and share them with the modeling community. The methodology proposed represents a paradigm shift in the burgeoning field of multiscale simulations and its application to solving complex clinical problems at the interface of engineering and biology. Predicting the progression of arterial thrombosis under circulation conditions, providing tools for improved pharmacological management as compared to existing empirics-based treatments, and providing a modeling tool for developing the next generation of devices with reduced thrombogenicity may lead to reduced mortality rates, improved patients' quality of life, and an overall reduction of the financial burden of the ensuing healthcare costs.